p. H.. CARPETSTTER 01!^ CRETACEOUS COMATTJLJE. 557 



possibly also in other species, they would seem to have extended 

 into the larval stem. Whether they remained open after the stem 

 was discarded, or whether their openings were closed up in the 

 interior of the centrodorsal, must of course remain uncertain. 



It seems to me very probable that the large and subdivided 

 internal cavities of the stems of Barycrinus and of the other genera 

 mentioned by Wachsmuth and Springer may have contained water 

 surrounding the central vascular axis, just as supposed above for 

 Cupressocrinus and for certain Comatulce. But these authors further 

 suppose that '^ there was ample communication with the surrounding 

 water " through the numerous branches at the base of the stem, all 

 of which are perforated. It is of course possible that the canals in 

 the rootlets of these large stems were not solely occupied by the 

 vessels, as is the case in the radicular cirrhi of Rhizocrinus ; but it 

 seems to me far less likely that these canals opened at the ends of 

 the rootlets, concealed as these were below the surface of the ooze. 

 The water entering (?) the stem by these passages would hardly 

 have been very useful for respiratory purposes. 



On the other hand it is not unlikely that the " large pores near 

 the base of the column, leading from without into the main cavity 

 directly through the walls," may have served to admit water into 

 the stem and thence into the coelom. The observations of Wachs- 

 muth and Springer that the complex perforated stems occur almost 

 exclusively in those forms which are destitute of hydrospires, or of 

 pores in the calyx, seem to point to this conclusion ; but there is 

 another possible mode of communication between the coelom and the 

 exterior, to which they do not refer. 



All recent Crinoids known to us have very minute ciliated pores 

 scattered over the perisome of the disk and the bases of the arms and 

 leading into the coelom. Miiller * has pointed out the correspondence 

 of this system of pores with the hydrospires of the Cystids (e. g. 

 Caryocrinus), and further research may reveal their presence in those 

 Palseocrinoids in which no hydrospires have yet been discovered. 

 So far as I know, they have not been looked for in these forms in the 

 same positions that they occupy in recent Crinoids, these parts being 

 rarely well preserved. In the recent species the pores are scattered 

 about on the surface of the disk and arm-bases in the neighbourhood 

 of the ambulacral grooves. When the perisome is plated as in Pen- 

 tacrinus and many Comatulce the water-pores occur on some of the 

 anambulacral plates which are close to the side plates of the grooves. 

 In the Palaeocrinoids, however, the surface of the body bearing the 

 grooves was not the external surface, as it was covered in by the 

 vault of rigid heavy plates. But there must have been a free ad- 

 mission of water beneath this vault by the so-called ambulacral 

 openings where the food-grooves of the arms extended over the disk 

 towards the subtegminal mouth. May not the areas of the body 

 between these grooves have been provided with ciliated pores leading 

 into the coelom, like those of recent Crinoids ? 



* "Ueber den Bau der Echinodermen," Abhandl. d. konigl, Akad. d. 

 Wissensch. zu Berlin, 1854, p. 64 (separate copy). 



